THE SUPREME COURT OF APPEAL OF SOUTH AFRICA
JUDGMENT
Reportable
Case no: 425/2022
In the matter between:
SIQALO FOODS (PTY) LTD APPELLANT
and
CLOVER SA (PTY) LTD RESPONDENT
Neutral citation: Siqalo Foods (Pty) Ltd v Clover SA (Pty) Ltd (425/2022) [2023]
ZASCA 82 (31 May 2023)
Coram: PONNAN, CARELSE and MATOJANE JJA and DAFFUE and
SIWENDU AJJA
Heard: 15 May 2023
Delivered: 31 May 2023
Summary: Agricultural Product Standards Act 119 of 1990 - s 3 read with
Regulations 2(1)(d) and (e) and 26(7)(a) - word or expression on a product label may
not be bigger than the class designation unless it is a registered trade mark or trade
name - s 6 read with Regulations 32(3) (a) and 32(4) - use of mark that conveys or
creates, or is likely to convey or crea te, a false or misleading impression as to the
quality, nature, class, origin or composition of a product.
2
___________________________________________________________________
ORDER
___________________________________________________________________
On appeal from: Gauteng Division of the High Court, Pretoria (Vuma AJ, sitting as
court of first instance):
The appeal is dismissed with costs, including those of two counsel.
___________________________________________________________________
JUDGMENT
___________________________________________________________________
Ponnan JA ( Carelse and Motojane JJA and Daffue and Siwendu AJJA
concurring):
[1] This appeal is concerned with the lawfulness of , inter alia, the names, words,
expressions or marks (the marks) depicted on the product container label (the product
label) of the appellant’s STORK BUTTER SPREAD (the product).
[2] The appellant, Siqalo Foods (Pty) Ltd (or more accurately its predecessor), was
incorporated and launched in 2018, when Remgro (Pty) Ltd completed its purchase of
Unilever South Africa (Pty) Ltd’s margarine ‘spreads’ business. The appellant’s
product line includes the STORK range consisting of, amongst others, the product.
[3] The respondent, Clover SA (Pty) Ltd, a wholly -owned subsidiary of the Clover
Group, is a manufacturer, marketer and purveyor of branded foods and beverages. It
manufactures and sells what is described as a modified butter product under its
registered trade mark BUTRO, which has been in use since July 1985.
[4] The compl aint of the respondent is that the appellant had commenced
promoting, marketing, distributing and selling the product as a butter product, when it
was in fact not such a product, but a modified butter product. The respondent objected
to such trade by the a ppellant on the basis that the product label misrepresents the
particular nature, substance, attributes, character and composition of the product,
thereby misrepresenting modified butter as a butter product.
3
[5] The respondent’s case is that the marks on the product label are proscribed in
at least two respects by the Agricultural Product Standards Act 119 of 1990 (the Act)
and the relevant regulations published thereunder (the regulations). 1 The first
regulates and prescribes, amongst others, the size of the marks that may be imprinted
on a product label or container (s 3 of the Act read with regulations 2(1)(d) and (e) and
26(7)(a)) and, the second prohibition forbids the use of any mark that conveys or
creates, or is likely to convey or create, a false or mi sleading impression as to the
nature, class or identity of a product (s 6 of the Act read with regulations 32(3)(a) and
32(4)).
[6] The respondent accordingly applied on 5 March 2021 to the Gauteng Division
of the High Court, Pretoria (the high court) to inte rdict and restrain the appellant from
competing unlawfully with the respondent and trading in contravention of the Act and
the Regulations. The application succeeded before Vuma AJ, who issued the following
order on 12 November 2021:
‘1. That the Respondent is interdicted and restrained from:-
1.1 competing unlawfully with the Applicant by using, selling, offering for sale, promoting,
advertising, delivering, marketing and/or in any way distributing for the purpose of sale,
modified butter products in a container and/or any other packaging and wrapping material
having a label imprinted thereon:-
1.1.1 as illustrated in the documents attached hereto as Annexures CF 2.1 – 2.6;
1.1.2 that is similar to the labels illustrated in Annexures CF 2.1 – 2.6;
1.1.3 in which the word “butter” appears as a dominant aspect or feature.
2. That the Respondent is interdicted and restrained from trading in contravention of
section 3 and 6 of the Agricultural Product Standards Act, 119 of 1990, as read w ith
Regulations 2, 3, 17, 18, 27 and 32 of the Regulations, GN R1510, published under that Act
in Government Gazette 42850, dated 22 November 2019, by using, selling, offering for sale,
promoting, advertising, delivering, marketing and/or in any way distri buting for the purposes
of sale, or offering for sale, modified butter products in a container and/or any other packaging
and wrapping material having an offending label imprinted thereon.
3. The Respondent is ordered, within 7 (SEVEN) days of this order, to: -
3.1. remove an offending label from all modified butter packaging and wrapping material,
and modified butter marketing and promotional material in their possession or under their
1 ‘Regulations Relating to the Classification, Packing and Marking of Dairy Products and Imitation Dairy
Products Intended for Sale in the Republic of South Africa, GN R1510, GG 42850, 22 November 2019.’
4
control; and
3.2. where an offending label is incapable of being removed from such material, to destroy
the material.
4. Costs of this application, including the cost consequent upon the employment of two
counsel are awarded to the applicant.’
The appeal is with the leave of the high court.
[7] It was common ground that the product is not butter, bu t something entirely
different – namely, a modified butter. Butter is a primary dairy product – a product
derived from or manufactured solely from milk. It may not contain any animal,
vegetable or marine fat.2 Butter is therefore ‘pure butter’. By contrast, modified butter
merely has the general appearance of butter, but it is not pure butter. It is an imitation
of butter. The product comprises a blend of 62% plant oils (fats) and other ingredients,
but only 38% primary dairy product.
[8] The product is as depicted here:
[9] Notably: (a) the word BUTTER is, when compared to all other words on the
label, dimensionally oversized and, therefore , visually accentuated; (b) the word
BUTTER is in bold blue font, capitalised and superimposed on a white letter-shadowed
background, further enhancing its visual accentuation; (c) the appellant’s goods mark,
2 See Regulation 17(1), which provides that butter and cultured butter with or without added foodstuff
shall be manufactured by churning or crystallisation of cream using the appropriate method; and not
contain any animal fat, vegetable fat or marine fat.
5
STORK, which appears above the word BUTTER, is dimensionally of a much smaller
size when compared to both the words BUTTER and SPREAD – it is, by contrast, not
superimposed on a white letter -shadowed background, thus, also further enhancing
the visual accentuation of th e word BUTTER on the product label; (d) the words
MODIFIED BUTTER, as it appears in the phrase MEDIUM FAT MODIFIED BUTTER
SPREAD WITH SUNFLOWER AND PALM OILS, are not in bold and are much smaller
dimensionally than the words BUTTER and SPREAD – that phrase appears in a
condensed and neutral script and is imposed on a dominant colourful graphic; and, (e)
the words WITH SUNFLOWER AND PALM OILS, which define the composition of the
product,3 are comparatively so small relative to the rest of the mark as to fade into
obscurity. The cumulative consequence of all of these considerations is that they serve
to under-accentuate the words MODIFIED BUTTER.
[10] The respondent accordingly contends that the product label, viewed as a whole,
not only contravenes the statutory labelling prohibitions, but also misrepresents or is
likely to create the misleading impression that the appellant’s modified butter product
is in fact a butter product. The ineluctable conclusion, so the contention proceeds, is
that the appellant’s product label was devised with exactly that misrepresentation and
misleading impression in mind.
[11] Five questions arise for consideration: First, do the marks on the product label
contravene the statutory labelling requirements? Second, does the appellant’s product
label misrepresent, or is it likely to misrepresent and create a misleading impression
regarding the respondent’s product, as a product of another nature, composition, class
or identity? Third, if the appellant is found to trade in contravention of these statutory
prohibitions, does that trade constitute unlawful competition? Fourth, did the
respondent have an alternative remedy available to it in lieu of this application? Fifth,
was the Minister of Agriculture, Land Reform and Rural Development (the Minister) a
necessary party and should she have been joined as a party to the suit?
As to the first:
[12] Sections 3(1)(a)(iii) and (v) of the Act empower the Minister to prohibit the sale
3 In other words, this phrase amounts to the product’s class designation.
6
of a product, unless the labelling of that product complies with the prescribed
requirements. In terms of regulation 2(1) (d), ‘[n]o person shall sell a dairy product or
an imitation dairy product in the Republic of South Africa . . . unless a container and
outer container in which such product is packed, is marked with particulars and in a
manner set out in regulations 26 to 31’. And, in terms of regulation 26(7)(a), ‘[n]o word
or expression [on a label] may be bigger than the class designation unless it is . . . a
registered trade mark or trade name’.
[13] The regulations define:
(i) ‘class designation’ as ‘the type of dairy product . . . as specified by these
regulations’;
(ii) ‘dairy product’ as ‘a primary dairy product, a composite dairy product or a
modified dairy product’;
(iii) ‘imitation dairy product’ as ‘any product other than a dairy product or a fat
spread, that is of animal or plant origin and in general appearance, present ation and
intended use corresponds to a dairy product’;
(iv) ‘primary dairy product’ as ‘milk or a product that has been derived or
manufactured solely from milk’; and
(v) ‘modified dairy product’ as ‘a product that, in so far as it relates to general
appearance, presentation and intended use, corresponds to a primary dairy product,
and of which not more than 50 per cent of the fat content, protein content and/or
carbohydrate content has respectively been obtained from a source other than a
primary dairy product.’
[14] It is not in dispute that the class designation of the appellant’s product is
‘modified butter’, not ‘butter’, and the words STORK BUTTER SPREAD and the words
BUTTER SPREAD, individually or compositely, are not registered trade marks. The
appellant attempts to suggest that, where regulation 26(7) (a) speaks of ‘a registered
trade mark or trade name’, it distinguishes between a registered trade mark, on the
one hand, and a trade name, on the other. It accordingly contends that STORK
BUTTER SPREAD is a trade name, not a registered trade mark.
[15] But, that contention is not supported by the evidence. First, the appellant had
applied to register STORK BUTTER SPREAD as a trade mark for use in respect of
7
certain goods, ie as a mark used ‘to indicate the origin of the goods in connection with
which it is used’. 4 Second, in order to obtain a clearance for its product label, the
appellant made a declaration to NejahMogul Technologies and Agric Services
(NejahMogul) (the assignee appointed in terms of the regulations as a product label
compliance inspector) that STORK BUTTER SPREAD is a trade mark. Third, in
response to the respondent’s demand issued before the launch of the application, the
appellant’s attorney recorded that the ‘complaint is directed at our clien t’s STORK
BUTTER SPREAD logo, which is a trade mark of our client. The trade mark complies
with sub-regulation 26(7)(a) of the Regulations’.
[16] Thereafter, the appellant expressed the view in its answering affidavit,
‘[w]hether or not STORK BUTTER SPREAD is a registered trade mark is . . . wholly
irrelevant’. Significantly, it did not state that STORK BUTTER SPREAD is its trade
name. Then, after the respondent’s replying affidavit had been filed, the appellant
proceeded (perhaps impermissibly so) to file a fo urth affidavit, described as a
‘supplementary answering affidavit’ in which it said, ‘[f]or the avoidance of all doubt, I
submit that “STORK BUTTER SPREAD” is utilised as a trade name’. But that
submission, being just that, lacked a proper factual foundation. And, if anything, it was
contradicted by what had gone before.
[17] It is in this context that the appellant argues (relying solely on the fourth affidavit
and completely ignoring what went before) that STORK BUTTER SPREAD is not a
trade mark but a trade name and, therefore, that it does not run afoul of regulation
26(7)(a). In that, the appellant should not be allowed to approbate and reprobate. Nor
should it be permitted that easily to escape the consequences of the representations
made by it. The appell ant’s volte-face ought, therefore, to be seen for what it is – a
belated attempt to escape the reach of s 3 of the Act, as read with regulations 2(1)(d)
and 26(7)(a).
As to the second:
4 Cowbell AG v ICS Holdings Ltd [2001] 4 All SA 242 (A); 2001 (3) SA 941 (SCA) para 10.
8
[18] Section 6 of the Act and regulations 32(3) (a)5 and 32(4)6 proscribe the use of
any mark that conveys or creates, or is likely to convey or create, a false or misleading
impression as to the nature, class or identity of a product.7
[19] Both parties accepted that the test to determine whether the use of the marks
would be likely to convey or create a false or misleading impression as to the nature,
class or identity of a product is the same as the test applied by the courts to determine
the likelihood of deception and/or confusion for the purposes of trade mark
infringement as well as passing off. Where a court considers such likelihood, it must
take ‘a commonsensical approach to the language on the labels . . . and to the visual
impressions created by them in order to resolve . . . the particular issue between the
parties’.8
[20] In the course of the determination of a likelihood of deception and/or confusion,
the commonsensical approach by a court is guided by the following principles. The
first impression is usually determinative of the issue. A court does not have to peer too
closely at the offending article to make the determination as to whether it is likely to
mislead.9 The court should notionally transport itself from the court room to the
marketplace and look at the article as it will be seen there by consumers. 10 When
considering the likelihood of deception and/or confusion, regard must be given to the
5 Regulation 32(3)(a) states that ‘no word, mark, illustration, depiction or other method of expression
that constitutes a misrepresentation or directly or by implication creates or may create a misleading
impression regarding the quality, nature, class, origin or composition of a diary product or an imitation
dairy product shall be marked on a container of such product’.
6 Regulation 32(4) provides that ‘no registered trade mark or brand name which may possibly, directly
or by implication, be misleading or create a false impression of the contents of a container or outer
container containing a dairy product or an imitation dairy product, shall appear on such a container ’.
7 Regulations 3 and 17 state that a product must be classified in accordance with specified classes and
that butter may not contain ‘any animal fat, vegetable fat or marine fat’. It must follow that a label that
represents ‘modified butter’ as ‘butter’ would also contravene the Regulations.
8 William Grant and Sons Ltd and Another v Cape Wine and Distillers Ltd and Others [1990] 4 All SA
490 (C); 1990 (3) SA 897 (C) at 913A-B (William Grant).
9 In the context of the determination of the likelihood of confusion and /or deception in the trade mark
infringement cont ext, which approach equally applies here, see: Puma AG Rudolf Dassler Sport v
Global Warming (Pty) Ltd [2009] ZASCA 89; [2010] 1 All SA 25 (SCA); 2010 (2) SA 600 (SCA) para 9:
‘. . .the question of the likelihood of confusion or deception is a matter of fi rst impression and that one
should not peer too closely at the registered mark and the alleged infringement to find similarities and
differences’. See also John Waddington Ltd v Arthur E Harris (Pty) Ltd [1968] 3 All SA 360 (T) ; 1968
(3) SA 405 (T) at 409D.
10 Laboratoire Lachartre SA v Armour -Dial Inc [1976] 3 All SA 88 (T) ; 1976 (2) SA 744 (T) at 746D;
Plascon-Evans Paints Ltd v Van Riebeeck Paints (Pty) Ltd [1984] 2 All SA 366 (A); 1984 (3) SA 623
(A) at 642D-F (Plascon-Evans).
9
role played by the dominant feature of the offending article,11 because consumers will
be focused on that feature and will not necessarily be alerted to the fine point s of
distinction or definition in order to clear up confusion.12 The confusion experienced by
consumers need also not be lasting – even if it lasts only for a ‘fraction of time’, it is
sufficient to find a likelihood of confusion, albeit that the confusion might later be
cleared up.13
[21] I daresay that this issue can be satisfactorily resolved without having regard to
the expert evidence adduced by the appellant, and thus without having to consider the
validity of the objection to its admissibility. This, perhaps because the court need really
go no further than take a commonsensical approach to the labels and to the visual
impressions created by them in order to resolve the issue, without expert assistance.14
However, because of the store sought to be set by the appellants on the evidence, it
may be as well to turn to a consideration of that evidence, namely the Melange
Concept Development Research Proposal (the Melange Proposal), Visual Attention
Service Shelf and Pack Testing Analysis (the VAS analysis) and Cat alyst Research
and Strategy Survey (the Survey). I do so , because as I shall endeavour to show in
what follows, despite the appellant’s argument to the contrary, it seems inescapable
from the evidence (such as it is) that the appellant may be ‘acting out a common
charade . . . sailing as close to the wind . . . without brewing up a storm of deception’.15
[22] In September 2019, the appellant commissioned qualitative consumer research
on the intended label for the product (the Melange Proposal). According to the
appellant, it then intended to use the sub -brand ‘Buttery’, instead of BUTTER
SPREAD. When consumers were exposed to the initial label:
‘. . . [T]hey felt that the sub -brand name “Buttery” was misleading and confusing – it implied
that the product is a “buttery-ish” product not actually a type of butter. Some consumers were
11 International Power Marketing (Pty) Ltd v Searles Industrials (Pty) Ltd 1983 (4) SA 163 (T) at 168H ;
Plascon-Evans fn 10 above at 641C-D.
12 Blue Lion Manuf acturing (Pty) Ltd v National Brands Ltd [2001] 4 All SA 235 (A); 2001 (3) SA 884
(SCA) para 3.
13 Orange Brand Services Ltd v Account Works Software (Pty) Ltd [2013] ZASCA 158 (SCA); 2013 BIP
313 (SCA) para 13; G C Webster, N S Page, C E Webster; G E Morley and I Joubert Webster and Page
South African Law of Trade Marks (2022) LexisNexis 4 ed SI 25 para 6.5.1 (Webster and Page).
14 William Grant fn 8 above at 912I-913B.
15 Red Bull GmbH v Rizo Investments (Pty) Ltd 2002 BIP 319 (T) at 323F (Red Bull).
10
even of the view that the sub-brand “Buttery” created the impression that the product could be
a margarine that has a butter flavour or texture.’
The initi al concept was therefore revised to exclude ‘Buttery’ and insert ‘BUTTER
SPREAD’ in its stead’. Further, so it was asserted:
‘Consumers unanimously preferred the revised version. This was because they were of the
view that the label clearly communicated the nature of the product – a butter spread.
It was also emphasised that the use of the product descriptor “butter spread” needed to be
bolder than normal on this new label, so as to clearly communicate the product contents and
differentiate this pack from other STORK products.’
[23] The Melange Proposal discloses the followi ng consumer reactions and
responses, namely that: (a) ‘butter spread’ is ‘a butter that is spreadable’; ‘a butter with
oil/more cream/something added to make it softer’; and ‘definitely a type of butter’; (b)
‘butter spread’ communicated that the product w as a spread made from butter as
opposed to a margarine spread; (c) the product descriptor ‘butter spread’ needed to
be bolder than normal so as to clearly communicate the product contents; and (d)
when consumers are asked if the product is butter – they say yes, it is pure butter but
with something added/removed to make it more spreadable.
[24] What is clear from these responses is that consumers believed that Stork butter
spread is ‘butter’ and ‘pure butter’. It is nowhere evident that any consumer had
correctly identified the product as ‘modified butter’, despite the fact that the design
label carried the phrase ‘medium fat modified butter spread’. But , that can only be
because that product class description is, as suggested earlier, essentially invisible.
This notwithstanding, the appellant accepted and implemented the Stork butter spread
format recommendation and, in the event, the product came to be launched in its
current packaging as ‘STORK BUTTER SPREAD’.
[25] In July 2020, Barrows Design and Marketing (Pty) Ltd (Barrows) received
instructions from the appellant to assess the effectiveness of the label, with particular
emphasis on what was described as the ‘stand -out’ of the label. According to the
evidence:
‘Barrows made use of Visual Attention Service (VAS), a web-based software tool that analyses
designs and photos and predicts 5 visual elements proven to attract human attention in the
11
first 3-5 seconds of viewing. It does this by using an algorithm to predict where people will
look.’
[26] The conclusion, am ongst others, was that the words ‘STORK BUTTER
SPREAD’ would have an 83% overall probability of determining and dictating the
visual fixation by a notional customer, over any other element of the label, ie t hat it
would be likely to ‘grab and hold’ the attention of a consumer in the first 3 to 5 seconds
of viewing. The further conclusion from the VAS Analysis was that the phrase
‘MEDIUM FAT MODIFIED BUTTER SPREAD’, being the very feature that should have
alerted the customer to the fact that the product is not pure butter but modified butter,
would have a ‘very low probability of attention’. That very low probability, may well
have been due to the significant under -accentuation of the class designation phrase
in the product label.
[27] Given the methodological s hortcomings, the appellant was constrained to
accept in its answering affidavit that ‘VAS is not proven research’. The obvious lack of
empirical cogency notwithstanding, the appellant then somewhat surprisingly added,
‘it is certainly an indicator to help guide design decisions . . .’. In those circumstances,
there is much to be said for the argument that no regard should be had to that
evidence. However, when the VAS Analysis, to the extent that reliance can be placed
on it, is read together with the Melange Proposal, it is difficult to resist the suggestion
that the product label was fashioned to focus consumer attention on the word ‘butter’.
[28] That leaves the Survey: As stated by Mr Michael Charnas, the Chief Executive
Officer of Catalyst Research and Strategy (Pty) Ltd (Catalyst):
‘4.1. Catalyst has been engaged by the [appellant’s attorney] . . . to conduct market
research to assist this honourable Court in determining, I am advised, the main issue in these
proceedings regarding whether the [appellant’s] STORK BUTTER SPREAD product will be
interpreted by consumers to be “pure” butter.
4.2. In order to test whether consumers will be confused into believing that the [appellant’s]
STORK BUTTER SPREAD product will be understood to be “pure” butter, the [appellant]
commissioned a market survey. This survey was titled the “Project Lard” survey and was
conducted during 12 to 17 March 2021.
. . .
Research method
12
4.4. The Project Lard Survey was conducted using an online panel survey and consisted
of a panel of 36,000 panelists, who opted to fill in the survey were utilised.
The universe
4.5. The universe represented by Project Lard Survey included a spread of demographics
across nine regions, all races, ages, genders and income levels.
4.6. However, the minimum earnings to a gross household had been restricted to an
income of R15,000 per month (who are amongst the LSM 8 – 10 income category) given that
this is calculated as the prime target market for butter and butter spread products.
Sample
4.7. In the survey, 914 respondents completed the survey. However, 459 respondents
qualified as butter and/or butter spread users in the LSM 8 – 10 income category.
Questionnaire
4.8. Qualifying respondents were shown a various range of (pure) butter and modified
butter products and asked to categorize each one as either:
4.8.1. Butter;
4.8.2. Butter spread;
4.8.3. Margarine;
4.8.4. None of the above;
4.8.5. Other – with a reason to be specified.’
[29] The Survey, so it would seem, purports to do exactly what the court has to
decide, namely whether a notional consumer would, or be likely to, confuse the
appellant’s imitation butter product with a pure butter prod uct. That aside, the
conclusion reached appears to contradict the conclusion in the Melange Proposal,
which found that consumers believed that the product is ‘butter’ or ‘pure butter’. For
the purposes of the determination of a likelihood of confusion by a notional consumer
in the notional marketplace, the methodology used in the Survey self -evidently fell
short. One searches in vain for questions that seek to recreate and emulate the
experience of a notional consumer in the notional marketplace.
[30] If anything, the Survey was ‘conducted under artificial conditions away from the
trade’, which renders the Survey ‘less probative’ and unreliable. 16 The participants
comprised an ‘online internet panel’, not actual, potential or even notional consumers
16 Webster and Page fn 13 above para 8.25, citing The European Limited v The Economist Newspapers
Limited [1996] FSR 431.
13
in a notiona l marketplace. They were required to complete an undisclosed internet -
based questionnaire, the content of which was also undisclosed, all of which makes it
impossible to properly assess both the efficacy and validity of the exercise. Labels
were provided t o the participants to study and consider before filling out the
questionnaire. The Survey, therefore, utterly ignored the notional marketplace and the
notional consumer, who would encounter the appellant’s product cheek by jowl with
other products.
[31] Somewhat alarmingly, the Survey, conducted during 12 to 17 March 2021,
claimed that a considerable number of participants knew the appellant’s product,
despite the fact that the product had only entered the market in limited volumes in
certain undisclosed locations on the 1 st of that month. How such product recognition
could occur in such a short period of time and with such limited market penetration, is
not explained.
[32] Significantly, the Survey also apparently pointedly avoided the real question
that ought to have been asked, namely, whether any survey participant could correctly
identify the appellant’s product as a modified or imitation butter. Nevertheless, it is
significant that the Survey concluded that a significant number of participants identified
and categorised the product as ‘butter’ and ‘butter spread’. Only a mere 5%
categorised the product as ‘margarine’ or ‘something else’. We are not told whether
that something else is ‘modified butter’. But the fact that 94% of the participants
identified the appellant’s product as butter and butter spread ought not to have come
as a surprise. It had been predicted in the Melange Proposal that if the butter product
descriptor on the label is made bolder so as to ‘grab and hold’ the consumer’s attention
within the first few seconds, consumers would perceive the product’s content as
‘butter’ or ‘pure butter’.
[33] It is so that although survey evidence is hearsay, it can be admitted as evidence
in terms of the Law of Evidence Amendment Act 45 of 1988.17 However, even if survey
opinions and conclusions are admitted into evidence, a court is at large to decline to
17 McDonald’s Corporation v Joburgers Drive -Inn Restaurant (Pty) Ltd and Another ; McDonald’s
Corporation v Dax Prop CC and Another ; McDonald’s Corporation v Joburgers Drive -Inn Restaurant
(Pty) Ltd and Another [1996] 4 All SA 1 (A); 1997 (1) SA 1 (A) at 26A-B.
14
accord the survey any evidentiary weight because the determination of a likelihood of
deception and confusion is within the exclusive province of the court, and no court
would surrender that determination to an expert. 18 The inherent deficiencies in the
Survey compounded, when read together with the articulated objectives, conclusions
and recommendations in the Melange Proposal and the VAS Analysis, may well
warrant rejection as unreliable. But, it is perhaps not necessary to go that far, because
to the extent that reliance can be placed on that evidence, they appear to warrant the
inference that the appellant had, in adopting the product label, acted out ‘a common
charade’.19
[34] The evidence, such as it is, appears to support the conclusion that the label
would, or at the very least would be likely to, convey or create a false or misleading
impression as to the nature, class or identity of the appe llant’s product. It also tends
to support the suggestion that the label was designed so as to mislead the public into
thinking that the product is a pure butter product.20
[35] As observed earlier, the word ‘butter’ is undeniably the dominant feature on the
appellant’s product label and , the product class designation ‘modified butter’ in the
phrase ‘medium fat modified butter spread with sunflower and palm oils’ , is virtually
invisible in the phrase (which itself is barely perceptible). It follows that the conclusion
that the label is likely to convey or create a false or misleading impression as to the
nature or class of the appellant’s product is inescapable – the peculiar get-up of the
label will self -evidently (or at least be likely to) deceive or conf use the notional
consumer into believing that the product is a butter product.
As to the third:
[36] It does not appear to be in dispute that if the appellant trades in contravention
of a statutory prohibition, such trade would also constitute an actionable wrong under
18 South African Human Rights Commission obo South African Jewish Board of Deputies v Masuku and
Another [2022] ZACC 5; 2022 (4) SA 1 (CC); 2022 (7) BCLR 850 (CC) ; 2022 (4) SA 1 (CC) para 144-
145; Gentiruco AG v Firestone SA (Pty) Ltd [1972] 1 All SA 201 (A); 1972 (1) SA 589 (A) at 616H.
19 Red Bull fn 15 above.
20 Ibid at 323F: ‘. . . the irresistible and only reasonable inference from a comparison of the marks . . .
is that [the] respondent is acting out a common charade . . . sailing as close to the wind . . . without
brewing up a storm of deception’.
15
the common law, namely unlawful competition 21 (which is actionable even if the
misrepresentation is innocent).22 On appeal, the appellant appears to have accepted
that if it is found to trade in contravention of the statutory prohibitions, then the
respondent has proven unlawful competition and that the court a quo was correct in
so finding.
As to the fourth:
[37] The appellant contends that the respondent had available to it an alternative
remedy under the Act which, so the contention goes, should have been pursued
instead of this application. On that score, s 3 of the Act, which states that the Minister
may ‘prohibit the sale of a prescribed product’, has been invoked.23 However, why it is
thought that s 3, which does no more than empower the Minister to take steps to
ensure compliance with the Act, would avail the respondent in the present
circumstances is far from clear. NejahMogul’s label clearance issued on 4 March 2020.
The argument that the respondent should have approached the Minister to prohibit the
sale of the product despite the grant of the clearance may well be untenable.
Moreover, the Act and the Regulations appear to make no provision for any such
further approach.
[38] In any event, s 3 of the Act falls far short of affording the respondent the remedy
sought in this application, namely to interdict and restrain the appellant’s continuing
unlawful conduct. As observed in Milestone Beverage CC and Others v The Scotch
21 Patz v Greene & Co 1907 TS 427; Pexmart CC and Others v H Mocke Construction (Pty) Ltd and
Another [2018] ZASCA 175; [2019] 1 All SA 335 (SCA); 2019 (3) SA 117 (SCA) paras 62 and 63(a);
Schultz v Butt [1986] 2 All SA 403 (A) ; 1986 (3) SA 667 (A) at 678F -H; Long John International Ltd v
Stellenbosch Wine Trust (Pty) Ltd 1990 (4) SA 136 (D) at 143G-I; Milestone Beverage CC and Others
v The Scotch Whisky Association and Others [2020] ZASCA 105; [2020] 4 All SA 335 (SCA); 2021 (2)
SA 413 (SCA) para 16 (Milestone).
22 Elida Gibbs (Pty) Ltd v Colgate-Palmolive (Pty) Ltd (1) [1988] 4 All SA 68 (W); 1988 (2) SA 350 (W)
at 358F-359A: ‘[w]here, however, a misstatement of fact relates to a fundamental or intrinsic quality of
the wares to be sold, thereby providing the advertiser with a competitive advantage, a plaintiff should
not be non-suited merely because the deception was innocent’.
23 Section 3(1) provides that the Minister may prohibit the sale of a prescribed product unless that
product is sold according to the prescribed class or grade; unless that product complies with the
prescribed standards regarding the quality thereof, or a class or grade thereof; unless the prescribed
requirements in connection with the management control system, packaging, marking and labelling of
that product are complied with; if that product contains a prescribed prohibited substance or does not
contain a prescribed substance; and unless that product is packed, marked and labelled in the
prescribed manner or with the prescribed particulars.
16
Whisky Association and Others 24 (Milestone) (citing with approval the judgment of
Trollip J in Johannesburg City Council v Knoetze and Sons25):
‘. . . [T]he purpose of an interdict is to restrain future or continuing breaches of a statute,
whereas the statutory r emedy of prosecuting and punishing an offender relates to past
breaches. Different considerations must therefore inevitably apply. For, while the statutory
remedies might be adequate to deal with past breaches, the civil remedy of an interdict might
be the only effective means of coping with future or continuing breaches.’26
[39] The respondent’s case is that the Act and the Regulations make no pro vision
for any form of relief even remotely similar to an interdict to restrain continuing unlawful
competition in the form of trade in contravention of a statutory prohibition. But, even if
there was a statutory remedy that could be invoked to address the unlawful
competition (and there appears to be none) , then applying the dictum in Milestone,
there is nothing that prevents the respondent from seeking an interdict in the high
court. Nothing, therefore, precluded the respondent from seeking the remedy of an
interdict for alleged trade in transgression of a statutory provision and, therefore,
unlawful competition in the court a quo.
As to the fifth:
[40] In support of the contention that the Minister has a direct and substantial
interest in the application and will be prejudicially affected by the relief sought, 27
reliance is sought to be placed on Esquire Electronics Ltd v Executive Video
(Esquire).28 However, as I see it, such reliance is facile. The dictum relied upon by the
appellant bears repeating:
24 Milestone fn 21 above.
25 Johannesburg City Council v Knoetze and Sons 1969 (2) SA 148 (W) at 150-155.
26 Milestone fn 21 above para 53.
27 As it was put in Judicial Service Commission and Another v Cape Bar Council and Another [2012]
ZASCA 115; [2013] 1 All SA 40 (SCA); 2013 (1) SA 170 (SCA) para 12:
‘It has by now become settled law that the joinder of a party is only required as a matter of necessity –
as opposed to a matter of convenience – if that party has a direct and substantial interest which may
be affected prejudicially by the judgment of the court in the proceedings concerned (see eg Bowring
NO v Vrededorp Properties CC and Another 2007 (5) SA 391 (SCA) para 21). The mere fact that a
party may have an interest in the outcome of the litigation does not warrant a non-joinder plea. The right
of a party to validly raise the objection that other parties should have been joined to the proceedings,
has thus been held to be a limited one (see eg Burger v Rand Water Board and Another 2007 (1) SA
30 (SCA) para 7; and Andries Charl Cilliers, Cheryl Loots and Hendrik Christoffel Nel Herbstein & Van
Winsen The Civil Practice of the High Courts of South Africa 5 ed vol 1 at 239 and the cases there
cited).’
28 Esquire Electronics Ltd v Executive Video [1986] 2 All SA 210 (A); 1986 (2) SA 576 (A).
17
‘In my opinion this point cannot properly be considered without the Registrar of Trade marks
having been joined. A decision upholding the point might have far -reaching consequences,
affecting the validity of numerous other trade marks in the register. More-over, Regulation 4(4)
provides that if any doubt arises as to what class any particular description of goods or services
belongs to, the doubt shall be resolved by the Registrar. It is plain that the Registrar is directly
and substantially interested in t he point and that it should not be decided unless he is given
an opportunity of being heard.’29
[41] The dictum underscores the fact that Esquire differs toto caelo from this case.
The point under discussion in that matter conceivably brought into question the validity
of numerous other registered trade marks.30 The judgment in Esquire was one in rem
in that it affects a public register. 31 Esquire, therefore, stated that the point could not
properly be considered without the R egistrar of Trade Marks (the Registrar) having
been joined, because a decision on that point would bind the Registrar and could have
far-reaching consequences for the validity of numerous other trade marks on the Trade
Marks Register. In contrast, the relief sought by the respondent in this application
affects and binds only the appellant and the respondent. It has no effect on, or
consequence for the Minister, the Department or any other party. For the present, the
relief sought also has no effect on, or i mplications for , the product label already
approved by NejahMogul.
[42] The role and powers of the Registrar also differ fundamentally from that of the
Minister. Contrasted with the Minister, the Registrar is a specialist in trade marks and
related law who has, in terms of s 45(1) of the Trade Marks Act 194 of 1993 (Trade
Marks Act), ‘all such powers and jurisdiction as are possessed by a single judge in a
civil action before the Transvaal Provincial Division of the Supreme Court’. The Trade
Marks Tribunal is furthermore a specialist tribunal especially created by the Trade
Marks Act to hear and decide all statutory issues relating to trade marks. The
provisions under consideration here do not create such a specialist tribunal to hear
and decide issues relating to matters falling within the scope of the Act. The Minister
and other functionaries, unlike the Registrar in Esquire, do not appear to have any
29 Ibid at 590I-591A.
30 Ibid at 590H.
31 The Gap Inc v Salt of the Earth Creations 2012 (5) SA 259 (SCA) para 2.
18
further role to play in this matter in relation to any of the orders sought by the
respondent. It follows that, like the other grounds of appeal raised by the appellant,
the non-joinder point must also fail.
[43] In the result, the appeal falls to be dismissed with costs, such costs to include
those of two counsel.
______________________
V M PONNAN
JUDGE OF APPEAL
19
Appearances
For the appellant: R Michau SC with L Harilal
Instructed by: Kisch Africa Inc, Pretoria
Phatshoane Henney Attorneys, Bloemfontein
For the respondent: AJ Bester SC with P Eilers
Instructed by: Hahn & Hahn Attorneys, Pretoria
McIntyre Van der Post Inc, Bloemfontein